Benchmark profile

MLVU mean average (MLVU (M-Avg))

A multi-task video understanding benchmark averaged across MLVU categories.

Data verified July 23, 2026

Benchmark score on MLVU (M-Avg) — July 23, 2026

BenchLM mirrors the published score view for MLVU (M-Avg). Qwen3.7 Plus leads the public snapshot at 87.4% , followed by Qwen3.6-27B (86.6%) and Qwen3.6-35B-A3B (86.2%). BenchLM does not use these results to rank models overall.

1Closed

Qwen3.7 Plus

Alibaba

qwen3-7-plus

87.4%

Overall 67.22Context 1M

2Open

Qwen3.6-27B

Alibaba

qwen3-6-27b

86.6%

Overall 53.82Context 262K

3Open

Qwen3.6-35B-A3B

Alibaba

qwen3-6-35b-a3b

86.2%

Overall 51.47Context 262K

3 modelsMultimodal & GroundedCurrentDisplay onlyUpdated July 23, 2026

Benchmark score table (3 models)

Score

Qwen3.7 PlusAlibaba · Closed

87.4%

Qwen3.6-27BAlibaba · Open weight

86.6%

Qwen3.6-35B-A3BAlibaba · Open weight

86.2%

The published MLVU (M-Avg) snapshot places Qwen3.7 Plus first at 87.4%. The third row is 1.2 points behind. The broader top-10 range is 1.2 points, so many of the published results sit in a relatively narrow band.

3 models have been evaluated on MLVU (M-Avg). The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. MLVU (M-Avg) is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About MLVU (M-Avg)

Year

2026

Tasks

General video understanding

Format

Video QA and understanding

Difficulty

Broad multimodal video reasoning

MLVU captures general-purpose video understanding rather than a single narrow skill. BenchLM tracks the mean-average summary row so provider comparison tables can be compared directly.

Qwen3.6 launch benchmarks

BenchLM freshness & provenance

Version

MLVU (M-Avg) 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does MLVU (M-Avg) measure?

A multi-task video understanding benchmark averaged across MLVU categories.

Which model scores highest on MLVU (M-Avg)?

Qwen3.7 Plus by Alibaba currently leads with a score of 87.4% on MLVU (M-Avg).

How many models are evaluated on MLVU (M-Avg)?

3 AI models have been evaluated on MLVU (M-Avg) on BenchLM.

Compare Top Models on MLVU (M-Avg)

Qwen3.7 Plus vs Qwen3.6-27B Qwen3.6-27B vs Qwen3.6-35B-A3B

Last updated: July 23, 2026 · BenchLM version MLVU (M-Avg) 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.